Blog List

Wednesday 9 November 2016

The Importance of Acacia Trees for Insectivorous Bats and Arthropods in the Arav

Published Date

Author 


Abstract


Anthropogenic habitat modification often has a profound negative impact on the flora and fauna of an ecosystem. In parts of the Middle East, ephemeral rivers (wadis) are characterised by stands of acacia trees. Green, flourishing assemblages of these trees are in decline in several countries, most likely due to human-induced water stress and habitat changes. We examined the importance of healthy acacia stands for bats and their arthropod prey in comparison to other natural and artificial habitats available in the Arava desert of Israel. We assessed bat activity and species richness through acoustic monitoring for entire nights and concurrently collected arthropods using light and pit traps. Dense green stands of acacia trees were the most important natural desert habitat for insectivorous bats. Irrigated gardens and parks in villages and fields of date palms had high arthropod levels but only village sites rivalled acacia trees in bat activity level. We confirmed up to 13 bat species around a single patch of acacia trees; one of the richest sites in any natural desert habitat in Israel. Some bat species utilised artificial sites; others were found almost exclusively in natural habitats. Two rare species (Barbastella leucomelas and Nycteris thebaica) were identified solely around acacia trees. We provide strong evidence that acacia trees are of unique importance to the community of insectivorous desert-dwelling bats, and that the health of the trees is crucial to their value as a foraging resource. Consequently, conservation efforts for acacia habitats, and in particular for the green more densely packed stands of trees, need to increase to protect this vital habitat for an entire community of protected bats.


Introduction


Desert habitats are resource limited by definition, putting flora and fauna under particular constraints [1]. Anthropogenic disturbance of such extreme natural habitats can have long-lasting deleterious effects [2]. Within mammals, bats are the second most species rich order [3], provide valuable ecosystem services [4], are abundant in many habitats, can easily be monitored through recordings of their powerful sonar vocalisations and are good bioindicators of habitat quality [5]. In desert areas of Israel (e.g. Negev, Arava and Judean) there are 17 species of insectivorous bats, representing more than half of the country’s desert mammals [6][7]. All insectivorous bats are protected by Israeli law and are either ‘vulnerable’, ‘near-threatened’ or ‘endangered’ on the International Union for Conservation of Nature (IUCN) red list for Israel [8].

Acacia trees are widely regarded as a keystone species with most desert fauna depending on them, either directly or indirectly, for food and shade [9][11]. They have an established positive impact on soil chemistry as nitrogen fixers [12] and increase herbaceous understory productivity [13]. Acacias hold crucial links to arthropods [14][16], such as ants, which live on acacias [17][20], bees which rely on acacia pollen [21] and bruchid beetles that infest seed pods [10][17][22][23]. Gazelle (Gazella dorcas), Arabian oryx (Oryx leucoryx), small nocturnal omnivorous rodents (Mastomys natalensis, Saccostomus campestris and Aethomys chrysophilus[24][25], ostriches (Struthio camelus) and giraffes (Giraffa camelopardalis[24]all consume the seeds of acacia, disperse and then fertilize pods aiding in germination while reducing the effect of seed parasites [23]. Three species, Acacia tortilis, A. raddiana, and A. pachyceras, provide the majority of wooded habitats in the Arava [26].

Acacia trees, particularly A. raddiana are in decline [27][28]; the total mortality of acacia trees in the Arava Rift Valley may be as high as 61% over 14 years [9][27]. This is primarily due to water stress, low recruitment of young acacia seedling and loss/change of habitat and water flow patterns [9][27]. As acacia trees rely predominantly on surface water, the latter factor is of great concern [29]. Additionally, there is a significant decline in annual precipitation, which is likely to have a negative effect on mortality and recruitment of acacias [30]. Rohner and Ward [23] predict that loss of acacia trees in the Middle East would lead to a significant loss of biodiversity in the region.Despite the wealth of research on the ecology of acacia trees it is almost completely unknown how and to what degree bats and their nocturnal arthropod prey might utilize acacia trees. Vaughan and Vaughan [31] found that the central African bat Lavia frons uses A. tortilis and occasionally A. elatior as a night roost from which to forage, and suggest that the bats are feeding on insects that are attracted to acacia trees. In Australia Vespadelus pumilus selectively roost in A. melanoxylon despite their relative rarity in the area [32]. Moreover, surveys of bats in the Sinai [33], Kenya [34] and Swaziland [35] found bat foraging activity at sites that contained acacia trees. None of these papers examined a specific interaction between bats and acacia trees, nor was there any explicit comparison to other available foraging habitats.

Here we examine activity levels and species richness of insectivorous desert bats and the abundance and richness of their arthropod prey in available natural and artificial desert habitats, including irrigated agricultural sites (date palms) and villages where desert-dwelling species are attracted to artificial light sources [6][36]. We hypothesise that acacias are a keystone genus in the nocturnal food web and therefore predict that nocturnal arthropods are diverse and abundant around acacia trees, and that bats are attracted to this foraging resource. Concurrently, we further hypothesise that the declining health of acacia habitats would negatively influence the bat and arthropod community and therefore predict that arthropod abundance and richness as well as bat activity and richness will be greater at dense green acacia stands than other available acacia habitats. Because artificial irrigation increases productivity in water limited ecosystems, we further predict that bat activity and arthropod abundance will be high in man-made habitats. We also predict that bat communities will differ between natural and man-made habitats, with a higher proportion of species that are typically recorded in the desert in natural habitats and more generalist synantropic species in the latter.

Materials and Methods

2.1 Study Site


The Arava rift valley, which connects the Dead Sea to the Red Sea, is an extremely arid desert with approximately 25–50 mm of annual rainfall and an average summer temperature of 31°C [37]. The area is characterised by ephemeral rivers that flood briefly after occasional, often distant, rains in most winters but otherwise remain dry (wadis). Scattered small settlements with irrigated parks, gardens and agricultural fields exist along the entire length of the valley. Potential foraging habitats for insectivorous bats in the Arava thus range from open desert with scarce vegetation that is typically dry in the summer months, through wadis with shrubs or trees to artificially irrigated and lit settlements and agricultural fields (e.g. date palms). The most stable biomass producers in the desert resource web are trees belonging to the genus Acacia. They occur in scattered lines along some wadis and, less frequently, in dense assemblages which are typically near small, often seasonal, springs, that are drying up due to aquifer pumping and climatic changes [38].

2.2 Habitats


We studied six habitat types; four natural: (1) densely packed green acacia trees (predominantly A. tortilis and A. raddiana) with trees clustered less than 30 m apart, (2) sparsely distributed green acacia trees with trees separated by greater than 50 m, (3) brown/barren acacia trees and (4) desert sites without acacia trees, as well as two modified by human habitation: (5) agriculture in the form of date plantations, and (6) irrigated vegetation at walkways or gardens in villages. We selected five different locations for each of the six habitat types giving a total of 30 sites in a 20×15 km area between the villages of Idan and Ein Yahav and using, from north to south, the accessible wadis Bitaron, En Zach, Masor, Shehaq and Dohan. There were only five dense green acacia stands in the research area; one in each of the five wadis. All other natural sites were selected to span the same north-south range around these (numbered 1–5 from north to south) preferentially with one site of each habitat within each respective wadi. Because adequate habitats were not always available in the same wadi, one barren acacia (B1), one sparse acacia (S5) and one no acacia (N5) site had to be located outside of the five wadis, and two barren acacia sites (B4 and B5) were in one wadi. There were five settlements in the area so artificial sites were selected in and around each settlement again spanning the area in the north-south range (Fig. 1). No sites were within 100 m of water available for drinking and most were >500 m away. All natural sites were pristine and away from public illumination, and town sites were the only artificial habitats that had any non-natural lighting. Potential roosts in the form of caves and crevices were plentiful throughout the study region; buildings, occupied or abandoned, were located within likely bat commuting distance of all sites (<5 km). We collected data at the 30 sites from April to August 2009. We visited sites randomly but made sure to visit one site for all six habitat types before starting with the next set of six different sites. After all 30 sites had been sampled once, we repeated this two more times but in a different random order. To minimise any potential lunar effect, we did not sample for five days around the full moon.

thumbnail
Figure 1. Satellite map of sites.
A is dense acacia stands, S is sparse acacia stands, B is barren acacia stands, N is non-acacia desert sites, V is village sites and D is date plantations. The five replicates of each habitat type are numbered one to five from north to south (Reprinted with permission from Esri, original copyright 2012).

2.3 Arthropod Sampling


We used a fluorescent light trap (Sylvania 15 W black light actinic bulb, 350 nm) suspended in front of a white cotton sheet to sample arthropods at each study site (for review of light trapping see [39]). Starting at 30 min after sunset, the light was turned on every hour for 30 minutes. All arthropods on the sheet were then collected during the following 10 minutes, and the light was then turned off for 20 minutes to avoid cumulative effects. We repeated this cycle four additional times and again once more beginning 90 min before sunrise. In villages, where ambient light might bias the attractiveness of our light trap, we placed the trap in darker areas, or those shielded from light. A pit trap was located less than 1 m from the sheet and checked every hour. We collected large arthropods in vials and smaller ones in pooters. This combination of methods has a bias against those arthropods not attracted to light but alternative methods were not suitable to the habitat/situation. Sweep netting was unviable as the net would get caught in acacia trees’ thorns damaging both the tree and the net while allowing all arthropods to escape; sticky traps quickly became covered in sand carried by the persistent winds; and, as many sites were in a national park, use of pesticides was not permitted. We identified all specimens at least to order and classified them as morphospecies. To create a reference collection of morphospecies hard-bodied specimens were collected, frozen and then pinned. Soft-bodied arthropods (Araneae, Scorpionidae and Solifugae) were stored in 70% ethanol. We analysed arthropod abundance per hour to account for periods of equipment failure in the light trap.

2.4 Acoustic Monitoring and Species Identification


At each site we used a full spectrum, direct recording automatic acoustic monitoring device with an omnidirectional microphone (BatCorder, EcoObs, Nuremberg, Germany) to record bat echolocation calls (@ 500 kHz and 16 bit) following the general approach outlined in Hayes et al. [40]. This device was hung from the edge of a tree 1–2 m from the ground. At sites where no trees were suitable a 1 m-high artificial stand was used. To avoid influencing recorded bat activity, the BatCorder was set at least 25 m from the arthropod trap. Once set, the BatCorder automatically records upon detection of a bat call and continues recording as long as bat calls are detected. After 800 ms of silence it ceases recording until triggered by a new call which starts a new file. Detection was assumed to be equal in all sites since desert habitats are all acoustically transmissible, with little canopy cover even at dense acacia sites. Within villages, sites were also open and there were no tall buildings obstructing bat flight. Agricultural sites were the most densely covered, but trees were still spaced 8–10 meters apart with crown diameters that leave gaps of 2–4 m between trees. There is currently no quantitative data on the transmissibility of habitat types for ultrasound, but large gaps in foliage in all habitats and the omnidirectional microphone of the BatCorder [41] mean that this is unlikely to have been a strong effect in this region. Due to differences in the source levels of different species’ echolocation calls, “loud” aerial hawkers are likely to be recorded over greater distances than “whispering” gleaning species [41]; this bias could not be eliminated but was equal across all sites. Activity was measured as number of bat passes per night and each recording file was defined as one pass. This is a conservative measure of bat activity if two passes of the same species are separated by less than 800 ms of silence. Multiple species present in the same file were defined as separate passes [42]. In a pilot survey in summer 2008, we identified the bat species foraging around acacia trees in the Arava using a combination of recordings from hand-released bats and descriptions of echolocation calls from studies in the broader region [33][43]. We established that desert bats in Israel can be identified to species level based on species/specific echolocation call design (see Fig. 2).

thumbnail
Figure 2. Spectrogram of one typical echolocation call from each identified species.
From left to right: Rhinopoma hardwickii, R. microphyllum, Nycteris thebaica, Asellia tridens, Rhinolophus hipposideros, R. clivosus, Pipistrellus kuhlii, Hypsugo bodenheimeri, Eptesicus bottae, Barbastella leucomelas, Otonycteris hemprichii, Plecotus christii and Tadarida teniotis. Spectrogram parameters: FFT 1024, frame 100%, overlap 98.43%, window flat top.

We used a weather monitoring device (Silva ADC Pro, Silva Sweden AB, Sweden) to record temperature, humidity and wind speed at the position of the BatCorder, with measurements taken one hour after sunset. At the beginning the night, wind was often blowing constantly with speeds of 5–10 km/h and occasionally as high as 20 km/hr. At a variable time, typically before midnight, this wind stopped abruptly and conditions remained calm for the rest of the night. Because of this pattern, wind was recorded as either present or absent one hour after sunset coinciding with the usual peak foraging activity. To standardise bat species identification and efficiently process the large number of recordings, we developed an automatic classification algorithm in SasLab Pro v. 4.40 (Avisoft Bioacoustics, Berlin, Germany). Peak frequencies at the start, end and maximum amplitude were measured for each echolocation call, which was then classified to species using defined frequency ranges per species. When compared to manual species identification the automatic classification correctly identified 95% of bat passes across all species (695 passes over 3 nights from pilot data in 2008); errors occurred when the recorded calls were too faint for the automatic classification to pick up (13 out of 115 passes for Rhinopoma hardwickii, 7 out of 537 passes for Hypsugo bodenheimeri), and when a H. bodenheimeri call overlapped with a R. hardwickii call it was mistakenly classified as Eptesicus bottae (5 passes); the passes of all other species were correctly identified in all cases. In order to reduce any further errors, some files had to be checked manually for potential misclassifications. This was necessary for recordings where (a) no call was classified/detected, or where (b) all calls classified as Otonycteris hemprichii, Plecotus christii or Tadarida teniotis, because low frequency noise was sometimes mistakenly classified one of these species. Echolocation calls from O. hemprichii and P. christii differ characteristically in the end frequencies, duration and the amount of spectral overlap between the first and second harmonic, but they had to be separated manually because automatic classification was unreliable. There were also rare misclassifications between solitary calls of three species with peak frequencies of approximately 30 kHz (E. bottae, R. hardwickii, and R. microphyllum). Hence, all files containing two or fewer such calls were also checked manually. Mist nets were routinely placed at each site, with the intention of confirming activity estimates, but due to the open nature of the desert habitats we rarely caught bats, stressing the advantages of acoustic monitoring for bat surveys. Bat captures and surveys were conducted under license #34615 given to CK by the Israel Nature and Park Authority, and all sites were visited with permission from land owners or the Israel Nature and Park Authority.

2.5 Statistical Analysis


Bat passes, arthropod abundance and the total number of bat species recorded at each site were heteroscedastic, therefore a log10(x+1) transformation was applied to the data enabling the use of parametric tests. A mixed-model analysis of variance (ANOVA) with one between-subjects factor (habitat) and one within-subjects factor (visit number) was performed on bat passes and arthropod abundance/hour. Green, dense acacia habitats were then individually compared to all other habitats using pairwise t-tests with sequential Bonferroni adjustment. Bat and arthropod species richness was measured as the number of species or morphospecies present at each site during the period of study. An ANOVA was performed on the number of bat species and arthropod morphospecies across habitats. A Pearson correlation test was performed to determine the relationship between bat passes and arthropod abundance. To test for any confounding effect of abiotic factors (temperature, humidity and wind) we performed a multivariate ANOVA for each visit cycle on the total number of bat passes per night and arthropod abundance per hour per night; we used a Bonferroni correction to control for multiple tests. All statistics were computed and graphs created using R-2.7.1 statistical environment (The R Foundation for Statistical Computing, 2008).

Results


We collected a total of 46,471 arthropods almost exclusively at the light trap over 533 hours with only 10 arthropods in pit traps. We identified 733 arthropod morphospecies in the following systematic groups: Lepidoptera (234), Coleoptera (100), Orthoptera (31), Mantoidea (12), Diptera (109), Neuroptera (30), Hymenoptera (39), Hemiptera (144), Blattaria (5), Odonata (4), Dermaptera (5), Isopoda (2), Ixodida (1), Pseudoscorpionida (1), Aranae (13), Solifugae (1) and Scorpiones (2).

Arthropod abundance was affected by both habitat type (F5,24 = 3.94; p  = 0.009; fig. 3a) and visit number (F2,48 = 5.26, p = 0.009; Fig. 3a). Arthropod abundance was lower at dense green acacia trees than at date sites (t21 = 3.42; p = 0.012), but after correcting for multiple testing there was no difference compared to other habitats (sparse acacia trees: t24 = 0.29, p = 0.77; barren acacia trees t27 = 2.55, p = 0.067; no acacia trees: t26 = 0.96, p = 0.69, village sites t24 = −2.19, p = 0.11). The number of arthropod morphospecies did not change significantly between habitat (F5,24 = 2.27, p = 0.079; Fig. 3c).

thumbnail
Figure 3. Arthropod abundance, bat activity and species richness for each habitat.
a: box plot of arthropod abundance per hour for three consecutive repeats (visits). b: box plot of total number of bat passes for three consecutive repeats (visits). c: total number of arthropod morphospecies (mean ± standard deviation. d: total number of bat species recorded (mean ± standard deviation).

We identified 13 bat species by their echolocation calls (Fig. 2): Rhinopoma hardwickii, R. microphyllum, Nycteris thebaica, Asellia tridens, Rhinolophus hipposideros, R. clivosus, Pipistrellus kuhlii, Hypsugo bodenheimeri, Eptesicus bottae, Barbastella leucomelas, Otonycteris hemprichii, Plecotus christii and Tadarida teniotis. We caught seven of these species in mist nets: R. hardwickii, A. tridens, R. clivosus, H. bodenheimeri, E. bottae, O. hemprichii and P. christii.

Over a total duration of 963 hours on 72 recording nights we recorded 6,575 bat passes in 5,586 files. Typically each file contained a single pass by a single individual, but passes of different species sometimes occurred in the same file indicating that multiple species were foraging at the same place and time. The number of bat passes was affected by both habitat type (F5,24 = 3.55; p  = 0.015; Fig. 3b) and visit number (F2,48 = 3.84, p = 0.028; Fig. 3b). The number of bat passes was significantly greater in dense green acacia trees than in all other natural desert habitats (sparse acacia trees t27 = 4.05, p = 0.001; barren acacia trees t20 = 5.88, p<0.001; no acacia trees: t28 = 4.23, p = 0.001) and date fields (t22 = 2.43, p = 0.05); but was not significantly different from village sites (t27 = 1.47, p = 0.15).

The number of bat species present was affected by habitat type (F5,24 = 2.80, p = 0.042; Fig. 3d), with dense green acacia trees having more species than barren acacia trees (t8 = 3.50, p = 0.04), but not significantly different from any other habitat (sparse acacia trees: t8 = 1.64, p = 0.44; no acacia trees: t8 = 5.54, p = 0.36; village sites: t8 = 6.40, p = 0.60 and date palms: t8 = 6.25, p = 0.75). Some species of bat were recorded almost exclusively in desert habitats and, within them, mostly in healthy acacia stands while others were more likely to be recorded in non-natural habitats (Fig. 4). For instance Rhinolophid species were recorded almost exclusively in natural habitats while P. kuhlii and T. teniotis were almost exclusively found in artificial sites. Other species (e.g. R. hardwickii and H. bodenheimeri) were more equally distributed between habitats.

thumbnail
Figure 4. The percentage of recorded bat passes in each habitat per species.
R.ha: Rhinopoma hardwickii, R.mi: R. microphyllum, A.tr: Asellia tridens, R.hi: Rhinolophus hipposideros, R.cl: R. clivosus, P.ku: Pipistrellus kuhlii, H.bo: Hypsugo bodenheimeri, E.bo: Eptesicus bottae, O.he: Otonycteris hemprichii, P.ch: Plecotus christii, T.te: Tadarida teniotis. Numbers in brackets indicate the total number of passes for that species.

There was a positive correlation between the number of bat passes and the arthropod abundance across all habitats but only 20.8% of the variation is accounted for by this relationship (R2 = 0.21; t88 = 4.81, p<0.001; Fig. 5). Wind had an effect on arthropod abundance in each visit cycle (1st visit: F1,28 = 12.73, p = 0.011; 2nd visit: F1,28 = 13.81, p = 0.008; 3rd visit: F1,28 = 9.20, p = 0.048) but not on bat activity (all visits F1,28<1.92, p>0.18). No other abiotic factor affected either arthropod abundance or bat activity (all F1,28<8.20, p>0.07).

thumbnail
Figure 5. The relationship between arthropod abundance and bat activity.
Each data point represents one entire night of sampling. Solid line is a linear regression (y = 0.627×−0.028; R2 = 0.21; t88 = 4.81, p<0.001).

Discussion


In accordance with our hypothesis, insectivorous bat activity was higher in dense green acacia stands than any other natural habitat, and species richness was high at habitats with dense green acacia trees. While dense green acacia trees only differed significantly from barren acacia trees in terms of bat species richness it was only at green acacia sites that we recorded all 13 species, which make up 76% of insectivorous bat species known from the deserts of Israel and Jordan [33][44][45]. Since natural sites were located along wadis, they may all be used as commuting routes. Thus the difference in bat activity levels, but not in species richness, between dense green acacias and the other natural desert habitats could be a result of bats flying through sparse acacia and no acacia sites en route to dense green acacias. These results indicate that healthy stands of acacia trees are key natural foraging resources for desert-dwelling insectivorous bats.

There was however only a weak link between habitat type and arthropods; no natural habitat differed in arthropod abundance or richness from green, dense acacias. One possible explanation for this is that our arthropod trapping method is biased towards light-attracted species, thus we are likely under sampling arthropods not attracted to light in all habitats. There might be a bias if these species’ abundance differed between habitats. While we attempted other sampling methods, these were not effective or not viable. The effect of visit number on both the number of bat passes and abundance of arthropods indicates that there is a potential seasonal component to habitat profitability and use. Healthy acacias remain green all year but partition flowering seasonally [46], and were in full flower during the 3rd visit. There is a general trend for arthropod abundance to decrease across the visits, particularly the 3rd visit, in the more barren sites (barren acacia trees and natural non-acacia sites). However, at green acacia trees (both dense and sparse) the level remains high during the 3rd visit in midsummer. It is therefore likely that green flowering acacias become even more important for bats and nocturnal arthropods as summer progresses and other habitats get less productive.

As we predicted, artificially irrigated and lit man-made habitats did have high arthropod abundance and bat activity. Date palms supported a greater abundance of arthropods than dense green acacias, while village sites and dense green acacias had equally high levels of bat activity. Moreover, arthropod and bat species richness for both date palms and village sites did not differ significantly from dense green acacias. These findings support the observation that for some species of bat, man-made habitats can, in fact, act as an alternative foraging resource [6].

Our results are consistent with previous studies that found bat activity correlated with arthropod abundance [47][48]. We recorded a range of bat species with different dietary niches [49], thus the activity of some species would likely correlate better with specific species of arthropods than others. Further studies into the diet of the bats in this area are needed to clarify this.

As predicted, some bat species relied more heavily on acacia trees than others; all species recorded here are listed at least as regionally vulnerable [8] (Table 1). Use of green acacia habitats was strongest in species typically recorded in deserts: R. clivosus, which mainly catches flying Coleoptera near vegetation, where echoes may come from objects that are not the target (cluttered environment) [6][44][49]; and O. hemprichii, which gleans terrestrial arthropods from surfaces [49][52] and tends to forage in xeric, sparsely vegetated, rocky environments [53] that are usually cluttered [6]. Conversely, P. kuhlii, a generalist in terms of prey and habitat selection that favours habitats with street lights [49][54], and T. teniotis, which hunts for flying insects in open spaces but is a generalist in terms of prey selection [6][33][49][54] were encountered mainly in non-natural habitats. Four species were recorded approximately equally in all habitats, both natural and artificial. All of these are aerial insectivores foraging in background cluttered space: H. bodenheimeri, is known to be a generalist in terms of both habitat and prey selection [6][49][55][56]R. hardwickii forages on aerial Coleoptera and swarming Hymenoptera in open habitats [49][57]R. microphyllumpredominantly consumes Coleoptera and is often found in sympatry with R. hardwickii [57][58]; and E. bottae, is a background cluttered space aerial insectivore [6]. Thus, artificial habitats created by the settlements appear to at least partially compensate for the habitat loss, but only for some of the desert species. Moreover, P. kuhlii, a species which has only expanded its range into desert areas of Israel following human habitation, could compete for resources with desert specialists in the vicinity of settlements [36][44][59][60].

thumbnail
Table 1. Regional and global conservation status of recorded bat species.

Three species were rarely recorded (Fig. 4): A. tridens and R. hipposideros tend to forage in highly cluttered environments [6][49][61][63], and P. christii is a recently isolated whispering bat presumed to consume mainly Lepidoptera [64]A. tridens and P. christii were sampled equally at healthy acacia and artificial sites while R. hipposideros was recorded mostly at healthy acacia sites and never in artificial habitats. Artificial light has been shown to negatively influence activity levels of R. hipposideros so it would not be expected in, or very near to, villages [65].

Of particular interest are two additional rare species that were only recorded outside our analysed sampling period, but exclusively at dense green acacia sites: B. leucomelas and N. thebaicaB. leucomelas has been caught only five times before in Israel. We have recorded it five times at three different dense green acacia tree sites. As they are so rare, nothing is known about habitat selection of B. leucomelas and this is the first occurrence of consistent recordings in the region. N. thebaica is a generalist/opportunistic feeder [49][66] found foraging in open savannah woodland areas [66]. It is also a whispering bat, hunting in flight or from a perch [66][67]. As these two as well as P. christii and O. hemprichii are presumed whispering bats with low intensity calls, they will have been under sampled and in fact be more prevalent in the area than determined by acoustic monitoring [68].

Many studies have found increased bat activity at sites with water [6][69][70], but that is not likely to be the explanation for the site-dependant differences we recorded. Sites of different habitats were all an approximately equal flight distance from standing water. Moreover, during the summer months the natural pools and springs often dry up completely, yet activity levels remained high.

Environmental factors are suggested to play a role in where bats forage. Temperature is negatively correlated with activity levels [71][76], while heavy rains can stop all foraging [71][72][77], and relative humidity is positively correlated to bat activity [78]. Environmental conditions remained similar during the period of each cycle of 30 sites, with a temperature range of less than ±5°C and without any precipitation. We found no significant effect of either temperature or humidity on either arthropod abundance or bat activity. The abiotic factor most likely to influence bat activity in the Arava is strong wind, because this increases energy expenditure for powered flight [79]. The effect of wind on bats is somewhat ambiguous, with evidence of both no change in bat activity [71] as well as a decrease in activity [54][78]. We did not find an effect of the presence of wind on bat activity over the whole night, but did on arthropod abundance, possibly because the wind moved the collection sheet thereby preventing insects from landing. Since wind was discontinuous and stopped abruptly during the night it is possible that bats shifted their activity to calm periods; thus leaving bat activity levels over the entire night unaffected.

Kunz et al. [4] reviewed the ecosystem services provided by bats, concluding that insectivorous bats potentially exercise a top down control of arthropods in both natural and agricultural ecosystems. The use of exclusion nets to determine the relative effects of predation by birds and bats on arthropods indicates that there is an equal or stronger effect of bats on arthropod abundance [80][82]. Moreover, bat predation of arthropods had an indirect effect on herbivory, providing a strong case for bats as biological agents of pest control [81][82]. Thus, the high activity level and diversity of insectivorous bats we found around dense healthy assemblages of acacia trees and also in irrigated agriculture might indicate that bats act as a biological control agent in both natural and agricultural habitats in the Arava.

Our findings provide evidence that acacia habitats are keystone foraging sites especially for rare bat species and desert specialists. Irrigated habitats in deserts are frequented by a selection of desert species, but synantropic species might increase resource competition [60]. We also give evidence that the health of the tree has a strong influence on activity level. Acacia trees’ further decline will have a significant impact on bats that forage in these areas. There is a need for better conservation and protection, particularly given the protected status of insectivorous bats in Israel and the ecosystem services they can provide.

Acknowledgments


Data was collected with the aid of Lauren Holt, Helen Hedworth, Orly Razgour, Shai Pilosof and especially Melia Nafus. Rangers from the Israel Nature and Park Authority were very helpful and friendly, particularly Yoram Hemo, Harel Ben Shahar, Roy Talbi and Asaf Tsoar as well as Eran Levin. We wish to thank Elizabeth Clare, Holger Goerlitz and two anonymous reviewers for comments on earlier versions of the manuscript. This is publication no. 789 of the Mitrani Department of Desert Ecology.

Author Contributions

Conceived and designed the experiments: TDH CK MWH. Performed the experiments: TDH. Analyzed the data: TDH MWH. Wrote the paper: TDH CK MWH.

References

  1. 1.Noy-Meir I (1973) Desert ecosystems: environment and producers. Annual Review of Ecology and Systematics 4: 25–51. doi: 10.1146/annurev.es.04.110173.000325 
  2. 2.Lovich JE, Bainbridge D (1999) Anthropogenic degradation of the southern California desert ecosystem and prospects for natural recovery and restoration. Environmental Management 24: 309–326. doi: 10.1007/s002679900235 
  3. 3.Simmons NB (2005) Order Chiroptera. In: Wilson DE, Reeder DM, editors. Mammal Species of the World: A Taxonomic and Geographic Reference. third ed: Johns Hopkins University Press. 312–529. 
  4. 4.Kunz TH, Braun de Torrez E, Bauer D, Lobova T, Fleming TH (2011) Ecosystem services provided by bats. Annals of the New York Academy of Sciences 1223: 1–38. doi: 10.1111/j.1749-6632.2011.06004.x 
  5. 5.Jones G, Jacobs DS, Kunz TH, Willig MR, Racey PA (2009) Carpe noctem: the importance of bats as bioindicators. Endangered Species Research 8: 93–115. doi: 10.3354/esr00182 
  6. 6.Korine C, Pinshow B (2004) Guild structure, foraging space use, and distribution in a community of insectivorous bats in the Negev Desert. Journal of Zoology 262: 187–196. doi: 10.1017/s0952836903004539 
  7. 7.Yom-Tov Y (1993) Character displacement among the insectivorous bats of the Dead Sea area. Journal of Zoology 230: 347–356. 
  8. 8.Dolev A, Perevolotsky A (2004) The Red Book: Vertebrates in Israel; Israel TINaPATSftPoNi, editor. 
  9. 9.Ward D, Rohner C (1997) Anthropogenic causes of high mortality and low recruitment in three acacia tree taxa in the Negev desert, Israel. Biodiversity and Conservation 6: 877–893. doi: 10.1023/b:bioc.0000010408.90955.48 
  10. 10.Or K, Ward D (2003) Three-way interactions between Acacia, large mammalian herbivores and bruchid beetles - a review. African Journal of Ecology 41: 257–265. doi: 10.1046/j.1365-2028.2003.00451.x 
  11. 11.Munzbergova Z, Ward D (2002) Acacia trees as keystone species in Negev desert ecosystems. Journal of Vegetation Science 13: 227–236. doi: 10.1111/j.1654-1103.2002.tb02043.x 
  12. 12.Belsky AJ, Amundson RG, Duxbury JM, Riha SJ, Ali AR, et al. (1989) The effects of trees on their physical, chemical, and biological environments in a semi-arid savanna in Kenya. Journal of Applied Ecology 26: 1005–1024. doi: 10.2307/2403708 
  13. 13.Weltzin JF, Coughenour MB (1990) Savanna tree influence on understory vegetation and soil nutrients in Northwestern Kenya. Journal of Vegetation Science 1: 325–334. doi: 10.2307/3235707 
  14. 14.Stone G, Willmer P, Nee S (1996) Daily partitioning of pollinators in an African acacia community. Proceedings of the Royal Society of London Series B-Biological Sciences 263: 1389–1393. doi: 10.1098/rspb.1996.0203 
  15. 15.Kruger O, McGavin GC (1998) Insect diversity of Acacia canopies in Mkomazi game reserve, north-east Tanzania. Ecography 21: 261–268. doi: 10.1111/j.1600-0587.1998.tb00563.x 
  16. 16.Kruger O, McGavin GC (2000) Macroecology of local insect communities. Acta Oecologica-International Journal of Ecology 21: 21–28. doi: 10.1016/s1146-609x(00)00112-0 
  17. 17.Ernst WHO, Tolsma DJ, Decelle JE (1989) Predation of seeds of Acacia tortilis by insects. Oikos 54: 294–300. doi: 10.2307/3565288 
  18. 18.Young TP, Stubblefield CH, Isbell LA (1997) Ants on swollen thorn acacias: Species coexistence in a simple system. Oecologia 109: 98–107. doi: 10.1007/s004420050063 
  19. 19.Kruger O, McGavin GC (1998) The influences of ants on the guild structure of Acaciainsect communities in Mkomazi Game Reserve, north-east Tanzania. African Journal of Ecology 36: 213–220. doi: 10.1046/j.1365-2028.1998.00130.x 
  20. 20.Palmer TM (2003) Spatial habitat heterogeneity influences competition and coexistence in an African Acacia ant guild. Ecology 84: 2843–2855. doi: 10.1890/02-0528 
  21. 21.Martins DJ (2004) Foraging patterns of managed honeybees and wild bee species in an arid African environment: ecology, biodiversity and competition. International Journal of Tropical Insect Science 24: 105–115. doi: 10.1079/ijt200411 
  22. 22.Hauser TP (1994) Germination, predation and dispersal of Acacia albida seeds. Oikos 70: 421–426. doi: 10.2307/3545781 
  23. 23.Rohner C, Ward D (1999) Large mammalian herbivores and the conservation of arid Acacia stands in the Middle East. Conservation Biology 13: 1162–1171. doi: 10.1046/j.1523-1739.1999.97300.x 
  24. 24.Miller MF (1995) Acacia seed survival, seed-germination and seedling growth following pod consumption by large herbivores and seed chewing by rodents. African Journal of Ecology 33: 194–210. doi: 10.1111/j.1365-2028.1995.tb00797.x 
  25. 25.Downs CT, McDonald PM, Brown K, Ward D (2003) Effects of acacia condensed tannins on urinary parameters, body mass, and diet choice of an acacia specialist rodent, Thallomys nicricauda. Journal of Chemical Ecology 29: 845–858. 
  26. 26.Horovitz A, Danin A (1983) Relatives of ornamental plants in the flora of Israel. Israel Journal of Botany 32: 75–95. 
  27. 27.Shrestha MK, Stock WD, Ward D, Golan-Goldhirsh A (2003) Water status of isolated Negev desert populations of Acacia raddiana with different mortality levels. Plant Ecology 168: 297–307. doi: 10.1023/a:1024431124954 
  28. 28.Ashkenazi S (1995) Acacia trees in the Negev, Israel, following reported large-scale mortality Jerusalem: Land Development Authority (HaKaren Ha Keyemet L'Israel). 
  29. 29.Sher AA, Wiegand K, Ward D (2010) Do Acacia and Tamarix trees compete for water in the Negev desert. Journal of Arid Environments 74: 338–343. doi: 10.1016/j.jaridenv.2009.09.007 
  30. 30.Ginat H, Shlomi Y, Batarshe S, Vogel J (2011) Reduction in precipitation levels in the Arava valley. Journal of Dead-Sea and Arava Research 1: 1–7. 
  31. 31.Vaughan TA, Vaughan RP (1986) Seasonality and the behavior of the African yellow-winged bat. Journal of Mammalogy 67: 91–102. doi: 10.2307/1381005 
  32. 32.Law BS, Anderson J (2000) Roost preferences and foraging ranges of the eastern forest bat Vespadelus pumilus under two disturbance histories in northern New South Wales, Australia. Austral Ecology 25: 352–367. doi: 10.1046/j.1442-9993.2000.01046.x 
  33. 33.Benda P, Dietz C, Andreas M, Hotovy J, Lucan RK, et al. (2008) Bats (Mammalia: Chiroptera) of the Eastern Mediterranean and Middle East. Part 6. Bats of Sinai (Egypt) with some taxonomic, ecological and echolocation data on that fauna. Acta Societas Zoologicae Bohemicae 72: 1–103. 
  34. 34.Webala PW, Oguge NO, Bekele A (2001) Bat species diversity and distribution in three vegetation communities of Meu National Park, Kenya. African Journal of Ecology 42: 171–179. doi: 10.1111/j.1365-2028.2004.00505.x 
  35. 35.Monadjem A, Reside A (2008) The influence of riparian vegetation on the distribution and abundance of bats in an African savanna. Acta Chiropterologica 10: 339–348. doi: 10.3161/150811008x414917 
  36. 36.Polak T, Korine C, Yair S, Holderied MW (2011) Differential effects of artificial lighting on flight and foraging behaviour of two sympatric bat species in a desert. Journal of Zoology 285: 21–27. doi: 10.1111/j.1469-7998.2011.00808.x 
  37. 37.Goldreich Y, Karni O (2001) Climate and precipitation regime in the Arava Valley, Israel. Israel Journal of Earth Sciences 50: 53–59. doi: 10.1092/1v61-fpgf-y5vk-adag 
  38. 38.Bruins HJ, Sherzer Z, Ginat H, Batarseh S (2012) Degredation of springs in the Arava Valley: anthropogenic and climatic factors. Land Degradation & Development 23: 365–383. doi: 10.1002/ldr.2149 
  39. 39.Young M (2005) Insects in flight; Leather SR, editor: Blackwell Science Publ, Osney Mead, Oxford Ox2 0el, Uk. 116–145 p. 
  40. 40.Hayes JP, Ober HK, Sherwin RE (2009) Survey and Monitoring of Bats. In: Kunz TH, Parsons S, editors. Ecological and Behavioral Methods for the Study of Bats. 2nd ed. Baltimore, MD USA: The Johns Hopkins University Press. 112–129. 
  41. 41.Adams AM, Jantzen MK, Hamilton RM, Fenton MB (2012) Do you hear what I hear? Implications of detector selection for acoustic monitoring of bats. Methods in Ecology and Evolution. 
  42. 42.Fenton MB (1970) A technique for monitoring bat activity with results obtained from different environments in southern Ontario. Canadian Journal of Zoology 48: 847–851. doi: 10.1139/z70-148 
  43. 43.Dietz C, von Helversen O (2004) Illustrated identification key to the bats of Europe. 1.0 ed. Germany: Tuebingen & Erlangen. 
  44. 44.Yom-Tov Y, Kadmon R (1998) Analysis of the distribution of insectivorous bats in Israel. Diversity and Distributions 4: 63–70. doi: 10.1046/j.1472-4642.1998.00012.x 
  45. 45.Shalmon B, Kofyan T, Hadad E (1993) A field guide to the land Mammals of Israel: their tracks and signs. Jerusalem, Israel: Keter Publishing House, Ltd. 
  46. 46.Stone GN, Willmer P, Rowe JA (1998) Partitioning of pollinators during flowering in an African Acacia community. Ecology 79: 2808–2827. doi: 10.1890/0012-9658(1998)079[2808:popdfi]2.0.co;2 
  47. 47.Anthony ELP, Stack MH, Kunz TH (1981) Night roosting and the nocturnal time budget of the little brown bat, Myotis lucifugus - effects of reproductive status, prey density, and environmental conditions. Oecologia 51: 151–156. doi: 10.1007/bf00540593 
  48. 48.Hayes JP (1997) Temporal variation in activity of bats and the design of echolocation-monitoring studies. Journal of Mammalogy 78: 514–524. doi: 10.2307/1382902 
  49. 49.Feldman R, Whitaker JOJ, Yom-Tov Y (2000) Dietary composition and habitat use in a desert insectivorous bat community in Israel. Acta Chiropterologica 2: 15–22. 
  50. 50.Holderied M, Korine C, Moritz T (2011) Hemprich's long-eared bat (Otonycteris hemprichii) as a predator of scorpions: whispering echolocation, passive gleaning and prey selection. Journal of Comparative Physiology A Neuroethology Sensory Neural and Behavioral Physiology 197: 425–433. doi: 10.1007/s00359-010-0608-3 
  51. 51.Arlettaz R, Dandliker G, Kasybekov E, Pillet JM, Rybin S, et al. (1995) Feeding-Habits of the Long-Eared Desert Bat, Otonycteris hemprichi (Chiroptera, Vespertilionidae). Journal of Mammalogy 76: 873–876. doi: 10.2307/1382757 
  52. 52.Fenton MB, Shalmon B, Makin D (1999) Roost switching, foraging behavior, and diet of the vespertilionid bat, Otonycteris hemprichii. Israel Journal of Zoology 45: 501–506. 
  53. 53.Gharaibeh BM, Qumsiyeh MB (1995) Otonycteris hemprichii. Mammalian Species 0: 1–4. doi: 10.2307/3504193 
  54. 54.Russo D, Jones G (2003) Use of foraging habitats by bats in a Mediterranean area determined by acoustic surveys: conservation implications. Ecography 26: 197–209. doi: 10.1034/j.1600-0587.2003.03422.x 
  55. 55.Riskin DK (2001) Pipistrellus bodenheimeri. Mammalian Species: American Society of Mammalogists. 1–3. 
  56. 56.Whitaker JO, Shalmon B, Kunz TH (1994) Food and Feeding-Habits of Insectivorous Bats from Israel. Zeitschrift Fur Saugetierkunde-International Journal of Mammalian Biology 59: 74–81. 
  57. 57.Whitaker JOJ, Yom-Tov Y (2002) The diet of some insectivorous bats from northern Israel. Mammalian Biology 67: 378–380. doi: 10.1078/1616-5047-00053 
  58. 58.Schlitter DA, Qumsiyeh MB (1996) Rhinopoma microphyllum. Mammalian Species: 1–5. 
  59. 59.Mendelssohn H, Yom-Tov Y (1999) Fauna Palaestina: Mammalia of Israel. Jerusalem, Israel: The Israel Academy of Science and Humanities. 
  60. 60.Razgour O, Korine C, Saltz D (2011) Does interspecific competition drive patterns of habitat use in desert bat communities? Oecologia 167: 493–502. doi: 10.1007/s00442-011-1995-z 
  61. 61.Bontadina F, Schofield H, Naef-Daenzer B (2002) Radio-tracking reveals that lesser horseshoe bats (Rhinolophus hipposideros) forage in woodland. Journal of Zoology 258: 281–290. doi: 10.1017/s0952836902001401 
  62. 62.Jones G, Rayner JMV (1989) Foraging Behavior and Echolocation of Wild Horseshoe Bats Rhinolophus ferrumequinum and Rhinolophus hipposideros (Chiroptera, Rhinolophidae). Behavioral Ecology and Sociobiology 25: 183–191. doi: 10.1007/bf00302917 
  63. 63.Zahn A, Holzhaider J, Kriner E, Maier A, Kayikcioglu A (2008) Foraging activity of Rhinolophus hipposideros on the island of Herrenchiemsee, Upper Bavaria. Mammalian Biology 73: 222–229. doi: 10.1016/j.mambio.2007.02.005 
  64. 64.Spitzenberger F, Strelkov PP, Winkler H, Haring E (2006) A preliminary revision of the genus Plecotus (Chiroptera, Vespertilionidae) based on genetic and morphological results. Zoologica Scripta 35: 187–230. doi: 10.1111/j.1463-6409.2006.00224.x 
  65. 65.Stone EL, Jones G, Harris S (2009) Street Lighting Disturbs Commuting Bats. Current Biology 19: 1123–1127. doi: 10.1016/j.cub.2009.05.058 
  66. 66.Gray PA, Fenton MB, Cakenberghe VV (1999) Nycteris thebaica. Mamalian Species: The American Society of Mammalogists. 1–8. 
  67. 67.Fenton MB, Gaudet CL, Leonard ML (1983) Feeding behaviour of the bats Nycteris grandis and Nycteris thebaica (Nycteridae) in captivity. Journal of Zoology 200: 347–354. doi: 10.1111/j.1469-7998.1983.tb02315.x 
  68. 68.Barclay RMR (1999) Bats are not birds - A cautionary note on using echolocation calls to identify bats: A comment. Journal of Mammalogy 80: 290–296. doi: 10.2307/1383229 
  69. 69.Russ JM, Montgomery WI (2002) Habitat associations of bats in Northern Ireland: implications for conservation. Biological Conservation 108: 49–58. doi: 10.1016/s0006-3207(02)00089-7 
  70. 70.Rebelo H, Brito JC (2006) Bat guild structure and habitat use in the Sahara desert. African Journal of Ecology 45: 228–230. doi: 10.1111/j.1365-2028.2006.00721.x 
  71. 71.Kunz TH (1973) Resource utilization - temporal and spatial components of bat activity in central Iowa. Journal of Mammalogy 54: 14–32. doi: 10.2307/1378869 
  72. 72.Rydell J (1991) Seasonal Use of Illuminated Areas by Foraging Northern Bats Eptesicus nilssoni. Holarctic Ecology 14: 203–207. doi: 10.1111/j.1600-0587.1991.tb00653.x 
  73. 73.O'Donnell CFJ (2000) Influence of season, habitat, temperature, and invertebrate availability on nocturnal activity of the New Zealand long-tailed bat (Chalinolobus tuberculatus). New Zealand Journal of Zoology 27: 207–221. doi: 10.1080/03014223.2000.9518228 
  74. 74.Meyer CFJ, Schwarz CJ, Fahr J (2004) Activity patterns and habitat preferences of insectivorous bats in a West African forest-savanna mosaic. Journal of Tropical Ecology 20: 397–407. doi: 10.1017/s0266467404001373 
  75. 75.Milne DJ, Fisher A, Rainey I, Pavey CR (2005) Temporal patterns of bats in the Top End of the Northern Territory, Australia. Journal of Mammalogy 86: 909–920. doi: 10.1644/1545-1542(2005)86[909:tpobit]2.0.co;2 
  76. 76.Ciechanowski M, Zajac T, Bitas A, Dunajski R (2007) Spatiotemporal variation in activity of bat species differing in hunting tactics: effects of weather, moonlight, food abundance, and structural clutter. Canadian Journal of Zoology-Revue Canadienne De Zoologie 85: 1249–1263. doi: 10.1139/z07-090 
  77. 77.Fenton MB, Boyle NGH, Harrison TM, Oxley DJ (1977) Activity patterns, habitat use, and prey selection by some African insectivorous bats. Biotropica 9: 73–85. doi: 10.2307/2387662 
  78. 78.Adam MD, Lacki MJ, Barnes TG (1994) Foraging areas and habitat use of the Virginia big-eared bat in Kentucky. Journal of Wildlife Management 58: 462–469. doi: 10.2307/3809317 
  79. 79.Schnitzler HU (1971) Bats in the wind tunnel. Zeitschrift fuer Vergleichende Physiologie 73: 209–221. doi: 10.1007/bf00304133 
  80. 80.Williams-Guillén K, Perfacto I, Vandermeer J (2008) Bats Limit Insects in a Neotropical Agroforestry System. Science 320: 70. doi: 10.1126/science.1152944 
  81. 81.Kalka MB, Smith AR, Kalko EKV (2008) Bats limit arthropods and herbivory in a tropical forest. Science 320: 71. doi: 10.1126/science.1153352 
  82. 82.Böhm SM, Wells K, Kalko EKV (2011) Top-down control of herbivory by birds and bats in the canopy of temperate broad-leaved okas (Quercus robur). PLoS ONE 6: 1–8. doi: 10.1371/journal.pone.0017857 
  83. 83.IUCN (2011) The IUCN Red List of Threatened Species. Version 2011.2. Available: http://www.iucnredlist.org. Downloaded 29 May 2012.

For further details log on website :
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0052999

Advantages and Disadvantages of Fasting for Runners

Author BY   ANDREA CESPEDES  Food is fuel, especially for serious runners who need a lot of energy. It may seem counterintuiti...